~0+. object in R programming

Question

I came across a solution for how to map non factorial data into a correlation matrix in which the solution uses an argument ~0+.. For reference, here's the code :

model.matrix(~0+., data=df) %>% 
  cor(use="pairwise.complete.obs") %>% 
  ggcorrplot(show.diag = F, type="lower", lab=TRUE, lab_size=2)

What does the object ~0+. do in this scenario?

I tried looking up the meaning of the operator ~ and tried to make sense of the formula but I cannot understand how it this formula helping us create a mapping of factors.

score 1 · Accepted Answer · answered Nov 03 '22 at 06:48

It is a formula object (see help("formula")for details).

. means all variables (columns), 0 means exclude the intercept.

You pass the formula to model.matrix, which creates a design matrix. It does automatic dummy encoding of factor variables (using treatment contrasts by default). Excluding the intercept ensures there is a column for each factor level in the design matrix.

~0+. object in R programming

1 Answers1